A hybrid predictive and prescriptive modelling framework for long-term mental healthcare workforce planning
Abstract
Over the past decade, severe staffing shortages in mental healthcare have worsened due to rising demand, further exacerbated by COVID-19. This demand is expected to grow over the next decade, necessitating proactive workforce planning to ensure sustainable service delivery. Despite its critical importance, the literature lacks a comprehensive model to address long-term workforce needs in mental healthcare. Additionally, our discussions with UK NHS mental health practitioners highlight the practical need for such a model. To bridge this gap, we propose a hybrid predictive-prescriptive modelling framework that integrates long-term probabilistic forecasting with an analytical stock-flow model for mental health workforce planning. Given the pivotal role of nurses, who comprise one-third of the mental health workforce, we focus on forecasting nursing headcount while ensuring the modelβs adaptability to broader healthcare workforce planning. Using statistical and machine learning methods with real-world NHS data, we first identify key factors influencing workforce variations, develop a long-term forecasting model, and integrate it into an analytical stock-flow framework for policy analysis. Our findings reveal the unsustainable trajectory of current staffing plans and highlight the ineffectiveness of blanket policies, emphasising the need for region-specific workforce strategies.